Discovering dark matter Di Subir Sarkar University of Oxford & - - PowerPoint PPT Presentation

Dark matter from aeV to ZeV: 3rd IBS-Multidark-IPPP workshop, Lumley Castle, 21-25 Nov 2016

Di Discovering dark matter

Subir Sarkar

University of Oxford & Niels Bohr Institute, Copenhagen

Snowmass CF1 WG summary, 1310.8327

1504.03198 1503.02641 1501.01200 1610.046111

Mass scale Particle Symmetry/ Quantum # Stability Production Abundance

ΛQCD

Nucleons

Baryon number τ > 1033 yr ‘freeze-out’ from thermal equilibrium

ΩB ~ 10-10

• cf. observed

ΩB ~ 0.05

We have a good theoretical explanation for why baryons are massive and stable

Durr et al, Science 322:2224,2008

Bethke, 1210.0325

We understand the dynamics (QCD) … and can even calculate the mass spectrum

‘Freeze-out’ occurs when annihilation rate: becomes comparable to the expansion rate where g ~ # relativistic species

Chemical equilibrium is maintained as long as annihilation rate exceeds the Hubble expansion rate

i.e. ‘freeze-out’ occurs at T ~ mN /45, with:

However the observed ratio is 109 times bigger for baryons, and there seem to be no antibaryons, so we must invoke an initial asymmetry:

Nucleons (predicted)➛ Nucleons (actual)➛

Zeldovich, 1965; Wolfram, 1979

Nucleons (predicted)➛ Nucleons (actual)➛

Why do we not call this the ‘baryon disaster’? cf. ‘WIMP miracle’!

Fields, Molaro & Sarkar, Review of Particle Properties, 2016

Although vastly overabundant compared to the natural expectation, baryons cannot close the universe (BBN ✜ CMB concordance)

… the dark matter must therefore be mainly non-baryonic

The SM allows B-number violation (through non-perturbative – ‘sphaleron-mediated’ – processes) … but CP-violation is too weak and SU(2)L x U(1)Y breaking is not a 1st order phase transition

Hence the generation of the observed matter-antimatter asymmetry requires new BSM physics … can be related to the observed neutrino masses if these arise from lepton number violation ➙ leptogenesis Ø B-number violation Ø CP violation Ø Departure for thermal equilibrium

‘See-saw’:

Any primordial lepton asymmetry (e.g. from out-of-equilibrium decays of the right-handed N) would be redistributed by B+L violating processes (which conserve B-L) amongst all fermions which couple to the electroweak anomaly – in particular baryons

An essential requirement is that neutrino mass must be Majorana … test by detecting neutrinoless double beta decay (and measuring the absolute neutrino mass scale)

Inverted hierarchy Normal hierarchy

Mass scale Particle Symmetry/ Quantum # Stability Production Abundanc e

ΛQCD Nucleons Baryon number τ > 1033 yr ‘freeze-out’ from thermal equilibrium Asymmetric baryogenesis ΩB ~10-10

• cf. observed

ΩB ~ 0.05 ΛFermi ~ GF-1/2 Neutralino? R-parity? Violated? (matter parity adequate to ensure B stability) ‘freeze-out’ from thermal equilibrium ΩLSP ~ 0.3

For (softly broken) supersymmetry we have the ‘WIMP miracle’:

Ωχh2 ⇥ 3 10−27cm−3s−1 ⇤σannv⌅T =Tf ⇥ 0.1 , since ⌅σannv⇧ ⇥

g4

χ

16π2m2

χ

⇤ 3 10−26cm3s−1

Leff ⊃ MAAµAµ + mf ¯ fLfR + m2

H|H|2

But why should a thermal relic have an abundance comparable to non-thermal relic baryons?

Mass scale Particle Symmetry/ Quantum # Stability Production Abundance

ΛQCD Nucleons Baryon number τ > 1033 yr ‘freeze-out’ from thermal equilibrium Asymmetric baryogenesis ΩB ~10-10

• cf. observed

ΩB ~ 0.05 ΛFermi ~ GF-1/2 Neutralino? R-parity? Violated? (matter parity adequate for p stability) ‘freeze-out’ from thermal equilibrium ΩLSP ~ 0.3

Ωχh2 ⇥ 3 10−27cm−3s−1 ⇤σannv⌅T =Tf ⇥ 0.1 , since ⌅σannv⇧ ⇥

g4

χ

16π2m2

χ

⇤ 3 10−26cm3s−1

But why should a thermal relic have an abundance comparable to non-thermal relic baryons?

Hidden sector (e.g. GMSB) matter also provides the ‘WIMPless miracle’ (Feng & Kumar, 0803.4196) … because: gh2/mh ~ gχ2/mχ ~ F/16π2M Such dark matter can have any mass: sub-GeV → ~few TeV

Mass scale Particle Symmetry/ Quantum # Stability Production Abundanc e ΛQCD ΛQCD’ ~ 6ΛQCD

Nucleons Dark baryon? Baryon number U(1)DB τ > 1033 yr (dim-6 OK) plausible ‘Freeze-out’ from thermal equilibrium Asymmetric baryogenesis (how?) Asymmetric (like the

• bserved baryons)

ΩB ~10-10 cf.

• bserved

ΩB ~ 0.05 ΩDB ~ 0.3

ΛFermi ~ GF-1/2

Neutralino? Technibaryon? R-parity (walking) Technicolour violated? τ ~ 1018 yr e+ excess? ‘Freeze-out’ from thermal equilibrium Asymmetric (like the

• bserved baryons)

ΩLSP ~ 0.3 ΩTB ~ 0.3

n0Χn0B 0.001 0.01 0.1 1 10 0.01 0.1 1 10 100 mΧTeV Χ

¢ ¢

➘ ΩTB/ΩB ≈ 6➚ So a O(TeV) mass technibaryon can be the dark matter … alternatively a ~few GeV mass ‘dark baryon’ in a hidden sector (e.g. into which the technibaryon decays) A new particle can naturally share in the B/L asymmetry if it couples to the W … linking dark to baryonic matter!

If they mix with the left-handed ‘active’ neutrinos then would behave as super-weakly interacting particles with an effective coupling: qGFermi

So they will be created when active neutrinos scatter, at a rate ∝ q2Gactive

Hence although they may never come into equilibrium, the relic abundance will be of order the dark matter for a mass of order KeV (however there is no natural motivation for such a mass scale) θ2

e,µ,τ ≡

|MDirac|2 |MMajorana|2 = Mactive Msterile ≈ 5 × 10−5 ✓Msterile KeV ◆−1

The SM admits a term which would lead to CP violation in strong interactions, hence an (unobserved) electric dipole moment for neutrons → requires θQCD < 10-10 θQCD must be made a dynamical parameter, by introducing a U(1)Peccei-Quinn symmetry which must be broken … the resulting (pseudo) Nambu-Goldstone boson is the QCD axion which acquires a small mass through its mixing with the pion: ma = mπ (fπ/fPQ)

+θQCDF ˜ F

When the temperature drops to LQCD the axion potential turns on and the coherent

• scillations of relic axions contain energy density that behaves like cold dark matter

with Ωah2~ 1011 GeV/fPQ … however the natural P-Q scale is probably Mstring~1018 GeV

Leff = F 2 + ¯ Ψ 6DΨ + ¯ ΨΨΦ + (DΦ)2 + Φ2

Hence QCD axion dark matter would need to be significantly diluted, i.e. its relic abundance is not predictable (or seek anthropic explanation for why θQCD is small?)

Javier Redondo

Many other possibilities for ‘axion-like particles’ … over a very large range of mass scales

Mass scale Lightest stable particle Symmetry/ Quantum # Stability ensured? Production Abundance

ΛQCD ΛQCD’ ~ 6ΛQCD

Nucleons Dark baryon? Baryon number U(1)DB τ > 1033 yr plausible ‘Freeze-out’ from equilibrium Asymmetric baryogenesis Asymmetric (like

• bserved baryons)

ΩB ~10-10 cf.

• bserved

ΩB ~ 0.05 ΩDB ~ 0.3

ΛFermi ~ GF-1/2

Neutralino?

Technibaryon? R-parity (walking) Techni- colour violated?

τ~1018 yr

‘freeze-out’ from equilibrium Asymmetric (like

• bserved baryons)

ΩLSP ~ 0.3 ΩTB ~ 0.3

Λhidden sector ~ (ΛFMP)1/2 Λsee-saw ~ΛFermi2/ΛB-L

Crypton? hidden valley? Neutrinos Discrete symmetry

(very model- dependent)

Lepton number τ ≳ 1018 yr Stable. Varying gravitational field during inflation Thermal (abundance ~ CMB photons) ΩX ~ 0.3? Ων > 0.003

Mstring /MPlanck

Kaluza-Klein states? Axions ? Peccei- Quinn ? Stable ? Field oscillations ? Ωa » 1!

Snowmass CF1 WG summary, 1310.8327

Several claims for putative signals have apparently been ruled out by more sensitive experiments … but are we making a fair comparison?

Di Direct detection ha has s fo focussed on

• n WIMPs,

s, so so is s most

• st se

sensi sitive at ~we weak scale

Th There are many ambiguities in interpreting the measured recoil rate:

★ Dark matter interacts differently with neutrons & protons (Giulani, hep-ph/0504157) if the mediator is a (new) vector boson … so e.g. the events seen by CDMS-Si can be consistent with the upper limits set by XENON100 or LUX ★ Then there are experimental uncertainties (instrumental backgrounds, efficiencies, energy resolution) + uncertainties in translating measured energies into recoil energies (channelling, quenching) + uncertain nuclear form factors …

No single experiment can either confirm or rule out dark matter

(and it is not a good strategy to look just under the WIMP lamp post!) ★ Moreover different experiments are sensitive to different regions of the (uncertain) dark matter velocity distribution, hence apparently inconsistent results (e.g. CoGeNT and DAMA) can be reconciled by departing from the assumed isotropic Maxwellian form (Fox et al, 1011.1915, Frandsen et al, 1111.0292, Del Nobile et al, 1306.5273)

Ma Many techniques for indirect detection … and many claims!

The PAMELA/AMS-02 anomaly (e+), WMAP/Planck ‘haze’ (radio), Fermi ‘bubbles’ + Galactic Centre ‘excess’ + 130 GeV line (γ-ray) … have all been ascribed to dark matter These are probes of dark matter elsewhere in the Galaxy so complement direct detection experiments … but we are just beginning to understand the astrophysical foregrounds!

These bounds require the scale Λ to exceed ~0.8 TeV, while perturbative unitarity requires gq, gχ < √4π i.e. mR < 2 TeV … so cannot rely on EFT description for higher energy collisions (Fox et al, 1203.1662)

q

¯ q

¯ χ

χ

‘Monojet’ events at colliders directly measure the coupling of dark matter to SM particles in an EFT, e.g. → → Recent move to ‘simplified models’ wherein the DM particle and its mediator to SM particles are specified to

• ptimise search strategies (1506.03116, 1607.06680)

The behaviour of dark matter associated with 4 bright cluster galaxies in the 10 kpc core of Abell 3827 Massey et al., 1504.03388 “The best-constrained offset is 1.62±0.48 kpc, where the 68% confidence limit includes both statistical error and systematic biases in mass modelling. […] With such a small physical separation, it is difficult to definitively rule out astrophysical effects operating exclusively in dense cluster core environments – but if interpreted solely as evidence for self-interacting dark matter, this offset implies a cross-section s/m=(1.7±0.7) x10-4 cm2/g (t/109yr)-2 where t is the infall duration.” The corrected value of the self-interaction cross-section is ~1.5 cm2/g (Kahlhoefer et al,

1308.3419, 1504.06576) … comparable to the upper limits derived from colliding galaxy clusters

q Searches for dark matter have focussed mainly on WIMPs so far but dark matter may be neither weakly interacting nor massive (and perhaps not even a particle)! q Lighter particles, which are just as well motivated, have just begun to be searched for with nuclear recoil experiments … complemented by collider searches for concommitant signals. q Dark matter may be coherent oscillations of axions necessitating very different search strategies (over a wide axion mass range). q Colliding galaxy clusters provide an interesting laboratory for strongly self-interacting dark matter (with the DM-stellar pop. separation predicted to be ~10-50 kpc for s/m ~ barn/GeV)

Interesting times ahead … recall that it took 48 years from the prediction of the Higgs boson to its discovery